]. Milk Food Technol, Vol. 37, No. 4 (1974) 199

MICROBIAL BETA-GALACTOSIDASE: A SURVEY FOR NEUTRAL pH OPTIMUM ENZYMES

L. C. BLANKENsHIP1 AND P. A. WELLS Dairy Foods Nutrition Laboratory, Nutrition Institute Agricultural Research Center, ARS, U. S. Departmant of Agriculture Beltsville, Maryland 20705 (Received for publication November 19, 1973)

ABSTRACT MATERIALS AND METHODS Downloaded from http://meridian.allenpress.com/jfp/article-pdf/37/4/199/2399667/0022-2747-37_4_199.pdf by guest on 02 October 2021 Pure cultures of yeast, molds, and bacteria were screened Cultures and media for neutral pH optimum .B-galactosidases ( lactases) that would One hundred twenty-five ·identified yeasts, molds, and bac­ be suitable in dairy l;lroducts applications. Only 2 of 125 teria were obtained from the culture collection of the North­ identified and 10 of 250 unidentified cultures warranted fur­ em Regional Research Center, Peoria, Illinois. Cultures of ther study. These cultures produced high levels of· .a-galacto­ the following genera were included: Penicillium, Aspergillus, sidase with moderate galactose product inhibition. Character­ Absidia, Cunninghamella, Mucor, Rhizopus, CircineUa, Blake­ ization of the partially purified enzymes from unidentified cul­ slea, Chlamydamucor, Streptomyces, Actinomyces, Actinopyc­ hwes revealed that all required either Na+, K+ or Mg++ nidium, Debaryomyces, Kluyveromyces, Pichia, Schwanniomy­ c::ation activation, were inhibited by Cu + +, Mn + +, and ces, Bullera, Brettanomyces, Candida, Cryptococcus, Rhoda­ Fe+ + +, were most active around pH 6.8, and were unstable torula, Torulopsis, Escherichia, and Bacillus. Additionally, during storage (at either - 196 C or 4 C} . except in the 250 unidentified organisms were isolated by enrichment cul­ presence of 0.5 M ammonium sulfate. Most of the enzymes ture techniques from soil samples taken from locations used compared favorably in performance with a commercially avail­ for dairy waste disposal. able .a-galactosidase when tested in skim milk. Yeasts were maintained on yeast extract-malt extract ( YM ) Renewed interest in enzymatic hydrolysis of lactose agar slants (5) having the following composition in grams per liter: yeast extract, 3.0; malt extract, 3.0; peptone, 5.0; in milk and dairy products has been stimulated by glucose, 10.0; and agar, 25.0. Molds were maintained on recent studies demonstrating the prevalence of lactose Czapek Dox agar (5). Bacteria were maintained on either intolerance among certain groups of consumers; (3, trypticase-soy agar (Baltimore Biological Lab) or yeast ex­ 7, 11). Furthermore, the pollution problem posed by b·act-peptone-lactose agar ( YPL) having the following com­ disposal of enormous quantities of cheese whey con­ position ,in grams per liter: yeast extract, 2.0; peptone, 10.0; ammonium sulfate, 4.0; lactose, 10.0; salts solution, 20.0 ml taining large amounts of lactose has forced food scien­ (2); and agar, 15. Cultures were stored at 2 C after in­ tists to seek more meaningful uses for whey. In­ cubation. creased use of whey in foods and feed, however, will necessarily be limited by consumer lactose tolerance Enrichment cultures and will consequently require some degree of lactose Enrichment cultures were grown in either YPL broth or in YM broth with lactose substituted for glucose. Each 1-2 g hydrolysis. of soil sample was inoculated into 200 ml of broth contained The feasibility of enzymatically hydrolyzing lactose in a 1-liter flask. Samples were incubated at 28 C. Samples in milk and dairy products has been reported by were periodically removed, diluted, and spread on appropri­ several laboratories (8, 14, 15). These studies have ate agar plates. Isolated colonies were picked and purified been carried out largely with yeast enzymes. Micro­ by isolation from additional spread plates. Isolates were identified as yeast, mold, or bacterium by gram strains and bial ,8-galactosidase, E.C. 3.2.1.22, (lactase) would phase contrast microscopy of wet mounts. appear to have the greatest commercial potential pri­ marily because of ease of production. Although a Screening for .a-galactosidase production broad variety of microorganisms are known to pro­ Two hundred-milliliter broth cultures contained in 1-liter duce ,8-galactosidase, (1, 9, 10, 13, 16, 17 ), very few flasks that were incubated for 24 to 48 h on a gyratory shaker have properties that would make them suitable for at 28 C were used for cell production during screening. The total mass of yeasts and bacteria was harvested by centri­ commercial use. Consequently we initiated a screen­ fugation, that of molds by filtration. Cells and mycelia were ing program to search for new sources of neutral pH washed once in 35 ml of 0.05 M potassium phosphate buffer optimum ,8-galactosidases that would be suitable for pH 6.8, and resuspended in fresh buffer to a 12-ml volume; dairy product applications. extracts were prepared by using a Sooifier cell disruptor, Model W 185 (Heat Systems - Ultrasonics, Inc., Plainview, N. Y. ). Sonication was done in an ice-water bath at 90-95 1Present address: Southeastern Regional Research Center, ARS, watts power setting. Particulate matter was removed by cen­ USDA, Athens, Georgia 30604. trifugation, and extracts were stored in liquid nitrogen. 200

Enzyme assay levels of e'Q.ZYIIle, and both were inhibited by galactose P-Galoctosidase activity of crude extracts was detennined in excess of 53%. These organisms, Klyveromyces by incubating the following reaction mixtll're for SO min at 35 C in screw cap tubes: enzyme; phosphate buffer pH 6.8, lactis NRRL Y-1118 and Klyveromyces fragilis NRRL 0.07 M; lactose, 0.139 M; and water in a total volume of 3.0 Y-1109, are known ,a-galactosidase producers and the ml. Parallel reactions were conducted in which 0.139 M D­ enzyme from Y-1109 has been well studied in dairy galactose was included to estimate product inhibition. Re­ products applications (15). actions were stopped by boiling 5 min. The amount of glu­ The remaining cultures producing substantial quan­ cose liberated was then determined according to the method described by Jasewicz (6). o-Nitrophenol-P-n-galactopyrano­ tities of ,B~galactosidase were isolated from enrich­ side ( ONPG) ( Calbiochem) was used as substrate for puri­ ment cultures. Crude extract specific activities, gal­ fication, thermal stability, pH optimum, activator, and in­ actose inhibition, enzyme yields, and morphological hibitor experiments. A unit of enzyme was defined as that type of the 10 most active organisms are given in amount which produced one .umole of glucose or 0-nitro­ Table 1. Generally, yeast enzymes were more sus­ phenol per minute under the reaction conditions specified above. Specific activity was defined as the number of units ceptible to galactose inhibition than were bacterial Downloaded from http://meridian.allenpress.com/jfp/article-pdf/37/4/199/2399667/0022-2747-37_4_199.pdf by guest on 02 October 2021 per mg protein. Protein was determined by the Biuret method enzymes. Since no attempt was made to determine (4). optimal cultural conditions, it is possible that higher Enzyme purification yields might have been obtained with other media. Selected cultures were grown in 2-liter quantities, and Failure to find any promising mold enzymes was not extracts of cells were prepared by using a French pressure surprising since most mold ,8-galactosidases have acid cell operated at 16,000 psi. Partial purification was done by pH optima and would not have been active at the fractional ammonium sulfate precipitation followed by DEAE screening pH. sephadex chromatography using a NaCl gradient elution in Although microbial ,8-galactosidases are generally pH 7.0, 0.05 M phosphate buffer. of intracellular origin, the advantages of processing Enzyme characterization an extracellular enzyme dictated screening of spent Thermal stability of partially purified enzymes was de­ broths for activity. No ,a-galactosidase activity was termined by heating microgram quantities of enzyme in 1.0 ml of water in screw cap tubes at the desired temperature for detected in any of the spent broths. 10 min. Samples were immediately cooled in an ice bath Microscopic examination of unidentified selected and the remainder of the reaction ingredients were added. cultures showed that all the bacteria were gram­ Reaction mixtures were then brought to 35 C, ONPG was negative rods. Yeasts were observed to be ellipsoidal added, and hydrolysis rates were determined from kinetic budding types. traces by using a recording Beckman DB spectrophotometer at 420 nm. Enzyme purification A variety of mono- and divalent cations were tested for Ammonium sulfate fractionation of selected culture activator or inhibitor effects on the selected, partially puri­ crude extracts generally resulted in precipitation of fied enzymes. These tests were conducted in tris-hydroxy­ methylamino methane chloride buffer pH 7.0 to avoid phos­ phate precipitates. TABLE 1. /1-GALACTOSIDASE PRODUCTION BY ENRICHMENT CULTURE ISOLATES Product inhibition studies with partially putified enzymes Unit!!/ were performed in which glucose or galactose was included Morpbologlcal Specific % Inhibition 200 ml in the standard reaction mixture at a final concentration of Culture type activity• by n-galactose culture 0.139 M. This corresponded to their concentration if total MTA-7 Bacterium 1.68 24.2 492 lactose hydrolysis had occurred in milk. MTB-28 Bacterium .476 29.3 116 pH optimum experiments were done in 0.05 M phosphate MTA-1 Bacterium 1.54 31.0 473 buffer over the range of 6 to 8. Results were normalized 105b Bacterium 3.40 36.6 1027 for differences in nitrophenol extinction coefficients at the Bel-17 Bacterium 1.91 13.2 600 various pH levels tested by reference to standard curves. Bel-15 Bacterium 1.22 29.1 411 MTB-18 Bacterium 1.38 38.0 224 REsULTS AND DISCUSSIOX 227 Yeast 2.14 51.6 534 217 Yeast 3.00 52.3 864 Screening 204 Yeast 3.71 49.5 682 The screening procedure used in this study was °Crude extracts. effective in selection of microorganisms producing high levels of ,a-galactosidase with moderate produe-1: the bulk of ,8-galactosidase in the 30-50% saturation inhibition and neutral pH as optima. Although many salt-cut. DEAE-sephadex chromatography of the of the identified organisms were from genera that had enzymes resulted in elution between the 0.30 and not been previously reported to produce ,a-galacto­ 0.35 M portion of the NaCl gradient in all instances. sidase, the hope for a ,a-galactosidase with unique Purifications varied among the enzymes ranging from properties did not materialize. Only two of the 8- to 15-fold relative to crude extract specific acti­ identified organisms surveyed produced substantial vities with recoveries ranging from 60 to 80%. MICROBIAL BETA-GALACI'OSJDASE 201

TABLE 2. THEIIMAL STABILITY OF SELECTED tf-GALACTOSIDASES actosida~es at various pH levels are given in Table 3. Percent All enzymes were most active at pH 6.8. Source of acttvtty remaining after 10 min heating at tl-galactosldase uc uc oc nc Influence of cations MTA-7 76.6 58.6 17.2 -" The influence of cations on the activity of fi-gal­ MTB-28 89.0 57.4 23.5 0 actosidases varies with the origin of the enzyme, the MTA-1 97.0 66.7 17.9 lOSt, 98.5 76.5 0 substrate, and the pH of the reaction (13). fi-Gal­ Bel-17 88.7 37.1 0 actosidases of bacterial origin in this survey were Bel-15 66.2 0 simffu.r to the Escherichia coli enzyme showing the MTB-18 79.3 18.5 greatest stumulation with Na + ( 0.1 M) and lesser 227 97.0 32.2 8.9 stimulation with K+ (0.1 M) and Mg++ (0.1 M). No 217 100 63.0 74.3 32.9 in 204 39.1 0 activity was observed in Tris-Cl buffer the absence of the above cations. Contrary to the findings of "Not determined. 4

Wallenfels (13 ), a low concentration of .M:n ++ ( 1Q- M) Downloaded from http://meridian.allenpress.com/jfp/article-pdf/37/4/199/2399667/0022-2747-37_4_199.pdf by guest on 02 October 2021 TABLE 3. EFFECT OF PH ON ONPG HYDROLYSIS BY or Fes+ ( 1Q-4 M) did not cause additional stimulation SELECTED tf-GALACTOSIDASES in the presence or absence of Na+. Instead, inhibi­ Specific activity• tion was observed. The source of enzymes or the Source of t~·&"alactosldase pll 6.0 pll 6.8 pll 8.0 assay buffer (Wallenfels used 0.04 M imidazole buffer MTA-7 6.6 11.0 7.1 pH 6.8 containing 0.04 M NaCl) might explain these MTB-28 2.0 6.6 4.0 MTA-1 6.6 27.6 20.0 observations.. Enzymes of yeast origin were activated lOeb 7.1 17.2 11.4 most by Mg+ + ( 0.1 M) and to a lesser extent by Na + Bel-17 5.3 21.0 1.0 (0.1 M) and K+ (0.1 M); this activation was similar Bel-15 5.8 ll.5 5.1 to that of other yeast enzymes (15). Mn++ and Fe+++ MTB-18 8.5 28.5 12.2 also inhibited the yeast enzymes. All the fi-galacto­ 227 2.1 4.0 0.7 sidases studied were greatly inhibited by Cu ++ ( 1o--• 217 4.8 8.5 5.2 2C4 16.7 38.1 23.8 M); this suggests that they are of the sulfhydryl type. Inhibition experiments were conducted in the pres­ "Specific activity number of units per rog protein. ence of the appropriate activating ion. A comparison of ONPG hydrolysi~ inhibition by Thermal stability glucose and galactose at concentrations equal to Most of the enzymes studied were essentially in­ total lactose hydrolysis in milk is shown in Table 4. activated by 10 min heating at 45 C (Table 2). The Generally, bacterial enzymes were inhibited by both enzyme from culture 217 was the exception retaining sugars except for MTA-1 which was not inhibited by 75% activity at 45 C. Commercial application of galactose. In contrast, yeast enzymes were not in­ ,8-galactosidase for neutral pH dairy products will hibited by glucose; in fact, cultures 217 and 227 were require good thermal stability. Ideally the enzyme stimulated to the extent of 28% and 47%, respectively. should function optimally either at 4 C and below Inhibition of lactose hydrolysis by glucose and trans­ or above 45 C to minimize the deleterious effects of galactosidation activity were not determined. microbial contaminant growth. The results for ,8- galactosidases in this study indicate that none would Comparison of lactases with commercial enzyme satisfactorily meet the high temperature specification. Shortly after initiation of this work a neutral pH Activities of our enzymes at 4 C or below were not optimum fi-galactosidase produced by K. lactis, and determined. marketed as "Maxilact" (Enzyme Development Corp., 2 New York, N. Y. ) became commercially available. Storage stability Consequently, it became necessary to compare the Storage stability is another commercially important performance of our enzymes to Maxilact in skim milk. enzyme characteristic. All of the partially purified Reaction mixtures were set-up containing 30 units of ,8-galactosidases in this study were unstable during enzyme (based on lactose hydrolysis), 5.0 ml double storage in dilute buffer at 4 C or frozen in liquid strength reconstituted dry skim milk, and water to a nitrogen. Storage stability could be greatly extended final volume of 10.0 ml. Reactions were conducted for at least one month by addition of ammonium sul­ at 35 C and 1-ml samples were taken at 1-h intervals fate ( 0.5 M). Although other methods of stabiliza­ tion were not tested it is possible that some may be 2Mention of a trademark or proprietary product does not applicable. constitute a guarantee of warranty of the product by the U. S. Department of Agriculture, and does not imply its ap­ pH optimum proval to the exclusion of other products that may also be The specific activities of partially purified fi-gal- suitable. 202 BLANKENSBIP AND WELLS

TABLE 4. INHIBITION OF ONPG HYDROLYSIS BY 0.189 M REFERENCES GLUCOSE AND GALACTOSE 1. Anema, P. J, 1964. Purification and some properties Percent lnh!b!tlon of ~-galactosidase of Bacillus substilis. Biochim. Biophys. Source of ~-B'alactosldase Glucose Galactose Acta 89:495-502. MTA-7 25.0 9.9 2. Cove, D. J, 1966. The induction and repression of MTB-28 28.2 85.5 nitrate reductase in the fungus AspergiUus nidulans. Biochim. MTA-1 84.0 0 Biophys. Acta 118:51-56. 1~ 31.1 41.8 3. Dairy Cotmcil Digest No. 6, 42 ( 1971) 31. Bel-17 27.3 32.8 4. Gornall, A. G., C. J, Bardawil, and M. M. David. 1949. Bel-15 46.2 35.6 Determination of serum protein by means of the biuret meth­ MTB-18 33.8 28.4 od. J. Bioi. Chem. 177:751-756. 227 0 20.0 5. Haynes, W. C., L. J, Wickerham, and C. W. Hesseltine. 217 0 32.5 1955. Maintenance of cultures of industrially important micro­ 204 0 48.3 organisms. Appl. Microbial. 3:361-368. 6. Jasewicz, L., and A. E. Wassennan. 1961. Quantitative Downloaded from http://meridian.allenpress.com/jfp/article-pdf/37/4/199/2399667/0022-2747-37_4_199.pdf by guest on 02 October 2021 TABLE 5. A COMPARISON OF SKIM MILK LACTOSE HYDROLYSIS determination of lactase. J. Dairy Sci. 44:393-400. BY UI\'IDE2><'"TIFIED ~-GALACTOSIDASES WITH A COMMERCIALLY 7. Ketchmer, N. 1972. Lactose and lactase. Sci. Amer. AVAlLABLE ENZYME 227:70-78. Source of ~-B'alactos!dase % Lactose hydrolyzed S h., 85 c 8. Kosikowski, F. V., and L. E. Wierzbicki. 1972. Lactose hydrolysis of raw and pasteurized milks by Saccharomyces MTA-7 50.5 lactis lat:1:ase. J, Dairy Sci. 56:146-148. MTB-28 45.8 9. McFeters, G. A., W. E. Sandine, and P. R. Elliker. MTA-1 49.2 1967. Purification and properties of Streptococcus lactis ~­ 105b 45.4 galactosidase. Bacterial. 93:914-919. MTB-18 49.6 J, 10. Pomeranz, Y. 1964. Lactase ({3-D-galactosidase). I. Bel-17 46.5 Occurrence and properties. Food Technol. 88:682-687. Bel-15 40.4 11. Rosensweig, N. S. 1969. Adult human milk tolerance 227 38.2 and lactase deficiency. A review. J, Dairy Sci. 52:585- 204 48.5 Maxila.ct 51.4 587. 12. Somogyi, M. 1952. Notes on sugar determinations. J, Bioi. Chem. 195:19-23. over a 3-h period. Reactions were stopped by boil­ 13. Wallenfels, K., and R. Weil. 1972. /~-Galactosidase, ing 5 min. Protein was removed by the Symogii p. 618 ·663. In P. D. Boyer ( ed) The enzymes, 3rd ed., method (12), and the amount of glucose released was Academic Press, New York, N. Y. detennined as previously described. The results after 14. Wendorff, W. L., C. H. Amundson, and N. F. Olson. 1970. The effect of heat treatment of milk upon the hydro­ 3-h incubation as shown in Table 5 indicated that lysability of lactose by the enzyme lactase. J, Milk Food most of the enzymes compared favorably with Maxi­ Technol. 33:377-379, lact. Although not shown in the table it should be 15. Wendorff, W. L., C. H. Amundson, N. F. Olson, and noted that the rate of hydrolysis was substantially J. C. Garver. 1971. Use of yeast ft-galactosidase in milk and slower after the first hour of incubation. milk products. J, Milk Food Techno}. 34:294-299. Commercial utilization of f3-galactosidase may be 16. Wendorff, W. L. and C. H. AmtmdS'On. 1971. Char· acterization of beta-galactosidase from Saccharoft!yces fragilis. facilitated by immobilizing the enzyme on insoluble J. Milk Food Techno!. 34:300-306. supports. The effects of immobilization on our en­ 17. Wierzbicki, L. E., and F. V. Kosikowski. 1972. Lac­ zymes are currently being evaluated and will be re­ tase potential of various microorganisms grown in whey, J, ported later. Dairy Sci. 56:26-32.